U.S. patent application number 12/921190 was filed with the patent office on 2011-01-20 for image sensing apparatus.
This patent application is currently assigned to CANON KABUSHIKI KAISHA. Invention is credited to Takayuki Hara, Saori Hoda.
Application Number | 20110013061 12/921190 |
Document ID | / |
Family ID | 41255171 |
Filed Date | 2011-01-20 |
United States Patent
Application |
20110013061 |
Kind Code |
A1 |
Hoda; Saori ; et
al. |
January 20, 2011 |
IMAGE SENSING APPARATUS
Abstract
An image sensing apparatus comprises an image sensing unit
including an image sensor having a plurality of image sensing
pixels and a plurality of focus detection pixels discretely
arranged between the plurality of image sensing pixels, a shift
detection unit that detects a shift amount between images obtained
by a plurality of image sensing operations, a combining unit that
combines signals of the focus detection pixels contained in the
images obtained by the plurality of image sensing operations on the
basis of the shift amount detected by the shift detection unit, and
a focus adjustment unit that adjusts a focus of the imaging lens by
using signals of the focus detection pixels before combined by the
combining unit, and signals of the focus detection pixels that are
combined by the combining unit.
Inventors: |
Hoda; Saori; (Kawasaki-shi,
JP) ; Hara; Takayuki; (Yokohama-shi, JP) |
Correspondence
Address: |
FITZPATRICK CELLA HARPER & SCINTO
1290 Avenue of the Americas
NEW YORK
NY
10104-3800
US
|
Assignee: |
CANON KABUSHIKI KAISHA
Tokyo
JP
|
Family ID: |
41255171 |
Appl. No.: |
12/921190 |
Filed: |
April 27, 2009 |
PCT Filed: |
April 27, 2009 |
PCT NO: |
PCT/JP2009/058655 |
371 Date: |
September 7, 2010 |
Current U.S.
Class: |
348/294 ;
348/E5.045; 348/E5.091 |
Current CPC
Class: |
H04N 5/23248 20130101;
G03B 13/36 20130101; H04N 5/23258 20130101; H04N 5/23254 20130101;
H04N 5/23212 20130101; G02B 7/34 20130101; G02B 7/38 20130101; H04N
5/232122 20180801; H04N 5/3696 20130101 |
Class at
Publication: |
348/294 ;
348/E05.091; 348/E05.045 |
International
Class: |
H04N 5/335 20060101
H04N005/335 |
Foreign Application Data
Date |
Code |
Application Number |
Apr 30, 2008 |
JP |
2008-119057 |
Claims
1. An image sensing apparatus comprising: an image sensing means
including an image sensor having a plurality of image sensing
pixels for photoelectrically converting an object image formed via
an imaging lens to generate an image signal, and a plurality of
focus detection pixels discretely arranged between the plurality of
image sensing pixels; a shift detection means for detecting a shift
amount between images obtained by a plurality of image sensing
operations using signals of the image sensing pixels contained in
image signals obtained by the plurality of image sensing operations
that are continuously obtained by said image sensing means and
shift in position from each other; a combining means for combining
signals of the focus detection pixels contained in the images
obtained by the plurality of image sensing operations on the basis
of the shift amount detected by said shift detection means; and a
focus adjustment means for adjusting a focus of the imaging lens by
using signals of the focus detection pixels before combined by said
combining means, and signals of the focus detection pixels that are
combined by said combining means.
2. The apparatus according to claim 1, further comprising a shift
means for shifting a position of the image sensor to obtain the
image signals by the plurality of image sensing operations that are
shifted in position from each other.
3. The apparatus according to claim 1, wherein said combining means
combines the signals of the focus detection pixels contained in the
images obtained by the plurality of image sensing operations by
averaging, based on the shift amount, signals of the focus
detection pixels contained in the images obtained by the plurality
of image sensing operations, and signals of neighboring focus
detection pixels.
4. A method of controlling an image sensing apparatus having an
image sensing means including an image sensor having a plurality of
image sensing pixels for photoelectrically converting an object
image formed via an imaging lens to generate an image signal, and a
plurality of focus detection pixels discretely arranged between the
plurality of image sensing pixels, the method comprising: a shift
detection step of detecting a shift amount between images obtained
by a plurality of image sensing operations using signals of the
image sensing pixels contained in image signals obtained by the
plurality of image sensing operations that are continuously
obtained by the image sensing means and shift in position from each
other; a combining step of combining signals of the focus detection
pixels contained in the images obtained by the plurality of image
sensing operations on the basis of the shift amount detected in the
shift detection step; and a focus adjustment step of adjusting a
focus of the imaging lens by using signals of the focus detection
pixels before combined in the combining step, and signals of the
focus detection pixels that are combined in the combining step.
Description
TECHNICAL FIELD
[0001] The present invention relates to an automatic focus
adjustment technique in an image sensing apparatus such as a
digital camera.
BACKGROUND ART
[0002] These days, image sensors in solid-state image sensing
apparatuses such as a digital still camera have abruptly increased
the number of pixels, and need to perform high-precision image
processing at high speeds.
[0003] To achieve high-precision automatic focus adjustment (to be
referred to as AF hereinafter) at high speeds, there has
conventionally been proposed a technique of arranging AF pixels (to
be referred to as focus detection pixels hereinafter) as some
pixels of an image sensor 601, as shown in FIGS. 7 and 8. As shown
in FIG. 7, the light receiving portion is divided to make pupil
positions symmetrical to each other. One light receiving portion is
defined as a focus detection pixel (A image) 501, and the other is
defined as a focus detection pixel (B image) 502.
[0004] As shown in FIG. 8, the focus detection pixels (A images)
501 and focus detection pixels (B images) 502 are scattered at a
certain ratio to all the pixels of the image sensor. Focus
detection is done by comparing the phase differences between focus
detection pixels (A images) 604 and focus detection pixels (B
images) 605 at the same horizontal position.
[0005] In the conventional arrangement, focus detection pixels are
arranged discretely, so an alias (aliasing) readily occurs and
greatly degrades the AF performance.
[0006] As a conventional technique for reducing aliases contained
in pixel information, the following proposals have been made.
[0007] Japanese Patent Laid-Open No. 9-65219 proposes a technique
of repeating the displacement of an image sensor by a shift means
to form an equivalently optical low-pass filter, thereby
effectively avoiding generation of aliasing of a video signal.
[0008] Japanese Patent Laid-Open No. 9-74524 proposes the following
technique. When moire is detected upon receiving an image of the
first resolution, an image of the second resolution increased by an
image shift means is generated. If no moire is detected, no moire
removal circuit operates, suppressing power consumption.
[0009] However, the proposals by Japanese Patent Laid-Open Nos.
9-65219 and 9-74524 are not aimed at reducing aliases of focus
detection pixels scattering in an image sensor. These references do
not mention a measure against camera shake. In general, detection
of a positional shift amount by matching calculation using an
alias-containing image tends to fail. Thus, conventional techniques
hardly reduce aliases of focus detection pixels.
DISCLOSURE OF INVENTION
[0010] The present invention has been made to overcome the
conventional drawbacks, and reduces aliases contained in focus
detection pixels and improve the AF performance.
[0011] According to the first aspect of the present invention,
there is provided an image sensing apparatus comprising an image
sensing means including an image sensor having a plurality of image
sensing pixels for photoelectrically converting an object image
formed via an imaging lens to generate an image signal, and a
plurality of focus detection pixels discretely arranged between the
plurality of image sensing pixels, a shift detection means for
detecting a shift amount between images obtained by a plurality of
image sensing operations using signals of the image sensing pixels
contained in image signals obtained by the plurality of image
sensing operations that are continuously obtained by the image
sensing means and shift in position from each other, a combining
means for combining signals of the focus detection pixels contained
in the images obtained by the plurality of image sensing operations
on the basis of the shift amount detected by the shift detection
means, and a focus adjustment means for adjusting a focus of the
imaging lens by using signals of the focus detection pixels before
combined by the combining means, and signals of the focus detection
pixels that are combined by the combining means.
[0012] According to the second aspect of the present invention,
there is provided a method of controlling an image sensing
apparatus having an image sensing means including an image sensor
having a plurality of image sensing pixels for photoelectrically
converting an object image formed via an imaging lens to generate
an image signal, and a plurality of focus detection pixels
discretely arranged between the plurality of image sensing pixels,
the method comprising a shift detection step of detecting a shift
amount between images obtained by a plurality of image sensing
operations using signals of the image sensing pixels contained in
image signals obtained by the plurality of image sensing operations
that are continuously obtained by the image sensing means and shift
in position from each other, a combining step of combining signals
of the focus detection pixels contained in the images obtained by
the plurality of image sensing operations on the basis of the shift
amount detected in the shift detection step, and a focus adjustment
step of adjusting a focus of the imaging lens by using signals of
the focus detection pixels before combined in the combining step,
and signals of the focus detection pixels that are combined in the
combining step.
[0013] Further features of the present invention will become
apparent from the following description of exemplary embodiments
with reference to the attached drawings.
BRIEF DESCRIPTION OF DRAWINGS
[0014] FIG. 1 is a block diagram showing the circuit arrangement of
an image sensing apparatus according to the first embodiment of the
present invention;
[0015] FIG. 2 is a view for explaining a state in which an image
sensed by the second image sensing operation shifts from an image
(reference image) sensed by the first image sensing operation;
[0016] FIG. 3 is a view showing a state in which shifted focus
detection pixels in the second image sensing operation are averaged
and added to the positions of focus detection pixels in the first
image sensing operation;
[0017] FIG. 4 is a view showing a state in which shifted focus
detection pixels in the second image sensing operation are
interpolated to the same horizontal positions as those of focus
detection pixels in the first image sensing operation;
[0018] FIG. 5 is a view for explaining an AF frame;
[0019] FIG. 6 is a block diagram showing the circuit arrangement of
an image sensing apparatus according to the second embodiment of
the present invention;
[0020] FIG. 7 is a view showing focus detection pixels whose pupil
positions are divided symmetrically; and
[0021] FIG. 8 is a view showing a state in which focus detection
pixels exist discretely on an image sensor.
BEST MODE FOR CARRYING OUT THE INVENTION
[0022] Embodiments of the present invention will be described in
detail below with reference to the accompanying drawings.
First Embodiment
[0023] FIG. 1 is a block diagram showing the circuit arrangement of
an image sensing apparatus according to the first embodiment of the
present invention.
[0024] In FIG. 1, a lens driving circuit 102 drives an imaging lens
101. A stop 103 adjusts the exposure. An image sensor 104
photoelectrically converts an optical signal. In the image sensor
104, some photoelectric conversion cells (image sensing pixels) are
replaced with focus detection pixels. The imaging lens 101 is
illustrated as one lens, but is generally formed from a plurality
of lenses or lens units. An actuator 120 shifts the position of the
image sensor 104. A sync signal generator (to be referred to as an
SSG hereinafter) 105 generates a horizontal sync signal HD and
vertical sync signal VD of fixed cycles. A timing generator (to be
referred to as a TG hereinafter) 106 generates a control signal
synchronized with HD and VD to drive the image sensor 104. An A/D
converter 107 converts an analog image signal output from the image
sensor 104 into a digital image signal. A signal processing circuit
109 performs color conversion processing and the like. A focus
detection pixel extraction circuit 122 extracts only data of focus
detection pixels from image data. A phase difference AF circuit 108
detects the phase difference between a focus detection pixel (A
image) and a focus detection pixel (B image) as shown in FIG. 7. A
shift amount detection circuit 119 detects the shift amount of an
image continuously captured (continuously obtained) by the second
image sensing operation from image sensing pixels other than focus
detection pixels. A focus detection pixel combining circuit 121
combines AF extracted pixels in a plurality of image sensing
operations to generate an AF extracted pixel in one image sensing
operation. A switching circuit 123 selects an output from the focus
detection pixel extraction circuit 122 or focus detection pixel
combining circuit 121, and outputs the selected output to the phase
difference AF circuit 108. A memory control circuit 110 interfaces
a DRAM 111. A resizing circuit 112 enlarges/reduces image data. A
system controller 113 determines the mode and parameter of each
circuit and controls the overall image sensing apparatus. A monitor
115 displays image data. A video modulation circuit 114 modulates
image data in order to display it on the monitor 115. A compression
circuit 116 compresses image data according to the JPEG compression
scheme or the like. A detachable medium card 118 records compressed
image data. A card control circuit 117 interfaces the medium card
118.
[0025] The operation of the image sensing apparatus having the
arrangement as shown in FIG. 1 will be explained.
[0026] The stop 103 adjusts the quantity of light having passed
through the imaging lens 101 from an object to an appropriate
one.
[0027] The TG 106 generates a timing signal and drives the image
sensor 104 so as to operate the image sensor 104 in synchronism
with HD and VD generated by the SSG 105. The image sensor 104
photoelectrically converts an object image formed via the imaging
lens 101 into an analog image signal. The A/D converter 107
converts the analog image signal into digital image data.
[0028] The focus detection pixel extraction circuit 122 extracts
only focus detection pixel data from image data, and outputs the
extracted data to the phase difference AF circuit 108 via the focus
detection pixel combining circuit 121 and switching circuit
123.
[0029] The phase difference AF circuit 108 compares the phase
differences between the A and B images of the focus detection pixel
data, and outputs phase difference information to the system
controller 113. In accordance with the phase difference
information, the system controller 113 notifies the lens driving
circuit 102 of the driving amount of a focus adjustment lens in the
imaging lens 101, thereby adjusting the focus.
[0030] The signal processing circuit 109 performs signal processing
such as color conversion for image data obtained from image sensing
pixels. The image data is temporarily written in the DRAM 111 via
the memory control circuit 110, and read out again to the shift
amount detection circuit 119.
[0031] Shift amount detection executed in the shift amount
detection circuit 119 will be explained.
[0032] As shown in FIG. 2, focus detection pixels (A images) 202
and focus detection pixels (B images) 203 are scattered in an image
sensing region 201 of the image sensor 104.
[0033] Assume that information on the positions of the focus
detection pixels (A images) 202 and focus detection pixels (B
images) 203 is obtained in the first image sensing operation.
[0034] Before executing the second image sensing operation, the
actuator 120 shifts the position of the image sensor 104 from the
position of the image sensing region 201 to that of an image
sensing region 204. Then, the positions of the focus detection
pixels also shift to positions represented by focus detection
pixels (A images) 205 and focus detection pixels (B images) 206.
Note that the image sensor is moved in the embodiment, but the lens
may also be moved.
[0035] If camera shake occurs in an image sensing operation, the
position of the image sensing region 204 moved by the actuator 120
shifts to that of an image sensing region 207. The positions of the
focus detection pixels also shift to positions represented by focus
detection pixels (A images) 208 and focus detection pixels (B
images) 209. In this state, the second image sensing operation is
done.
[0036] As a method of obtaining the shift amount between images
obtained by the first and second image sensing operations, the
image sensed by the first image sensing operation, and the image
sensed by the second image sensing operation that contains a shift
generated by camera shake undergo matching calculation using image
sensing pixels other than focus detection pixels. The matching
calculation employs the following known method. More specifically,
image data in the first and second image sensing operations which
have undergone signal processing by the signal processing circuit
109 and are written in the DRAM 111 are read out again from the
DRAM 111. Then, a position where the differences between image
sensing pixels in the first and second image sensing operations
minimize as a whole is detected.
[0037] Shift-amount-corrected image data of two image sensing
operations that are obtained by matching calculation in the shift
amount detection circuit 119 are output to the focus detection
pixel combining circuit 121.
[0038] The focus detection pixel combining circuit 121 executes
combining processing by either of the following two methods.
[0039] <First Focus Detection Pixel Combining Method>
[0040] FIG. 3 shows the first focus detection pixel combining
method.
[0041] Four pairs of focus detection pixels (A images) 208 and
focus detection pixels (B images) 209 positioned around a given
focus detection pixel (A image) 202 and focus detection pixel (B
image) 203 undergo weighting calculation to superpose them on the
positions of the focus detection pixel (A image) 202 and focus
detection pixel (B image) 203. Then, averaging is executed. The
weighing coefficient is calculated from a shift amount obtained by
matching calculation in the shift amount detection circuit 119.
[0042] As shown in FIG. 3, AF2_A1, AF2_A2, AF2_A3, and AF2_A4
represent the values of four focus detection pixels (A images) 208
which are obtained by the second image sensing operation and are
positioned near focus detection pixels in the first image sensing
operation in which the value of the focus detection pixel (A image)
202 is AF1_A and that of the focus detection pixel (B image) 203 is
AF1_B. Also, AF2_B1, AF2_B2, AF2_B3, and AF2_B4 represent the
values of four focus detection pixels (B images) 209.
[0043] As shown in FIG. 3, H1 and H2 represent horizontal distance
ratios from AF1_A and AF1_B to four pairs of surrounding focus
detection pixels. Similarly, V1 and V2 represent vertical distance
ratios.
[0044] H1, H2, V1, and V2 satisfy H1+H2=1 and V1+V2=1,
respectively.
[0045] The values AF2_A and AF2_B from the four pairs of
neighboring focus detection pixels to the positions AF1_A and AF1_B
are calculated by
AF2.sub.--A=(AF2.sub.--A1*V2+AF2.sub.--A3*V1)*H2+(AF2.sub.--A2*V2+AF2.su-
b.--A4*V1)*H1 (1)
AF2.sub.--B=(AF2.sub.--B1*V2+AF2.sub.--B3*V1)*H2+(AF2.sub.--B2*V2+AF2.su-
b.--B4*V1)*H1 (2)
[0046] AF1_A and AF1_B, and AF2_A and AF2_B are averaged,
respectively. Pieces of focus detection pixel information AF_A and
focus detection pixel information AF_B at the obtained positions
are calculated:
AF.sub.--A=(AF1.sub.--A+AF2.sub.--A)/2 (3)
AF.sub.--B=(AF1.sub.--B+AF2.sub.--B)/2 (4)
[0047] Similarly, AF extracted images obtained by a plurality of
image sensing operations are combined. The values of AF extracted
images during combination are stored in the focus detection pixel
combining circuit 121.
[0048] In the above-described way, the focus detection pixel
combining circuit 121 can increase the number of focus detection
pixels, thereby increasing the resolution and reducing aliases. In
this case, increasing the resolution means increasing the number of
SAF pixels when calculating the defocus amount.
[0049] Information of a focus detection pixel (A image) 205 and
focus detection pixel (B image) 206 obtained by averaging images
sensed by a plurality of image sensing operations is output to the
switching circuit 123, and then to the phase difference AF circuit
108.
[0050] The phase difference AF circuit 108 obtains the defocus
amount by comparing the phase differences between A and B images
using the focus detection pixels (A images) 202 and focus detection
pixels (B images) 203 before combination, and the focus detection
pixel (A image) AF_A and focus detection pixel (B image) AF_B. The
phase difference AF circuit 108 outputs the obtained defocus amount
as focusing information to the system controller 113.
[0051] <Second Focus Detection Pixel Combining Method>
[0052] FIG. 4 shows the second focus detection pixel combining
method.
[0053] Assume that AF information of positions indicated by focus
detection pixels (A images) 202 and focus detection pixels (B
images) 203 is obtained in the first image sensing operation, and
AF information of positions indicated by focus detection pixels (A
images) 208 and focus detection pixels (B images) 209 is obtained
in the second image sensing operation.
[0054] Interpolation is done by weighting at positions where no AF
information is obtained by the first and second image sensing
operations, thereby increasing the sampling points of focus
detection pixels.
[0055] At the positions of combined focus detection pixels (A
images) 401 and combined focus detection pixels (B images) 402 in
FIG. 4, A images are interpolated by A images, and B images are
interpolated by B images.
[0056] The focus detection pixel value AF_A of the combined focus
detection pixel (A image) 401 is calculated using the focus
detection pixel values AF1_A1, AF1_A2, AF2_A1, and AF2_A2 of
neighboring focus detection pixels (A images):
AF.sub.--A={(AF1.sub.--A1*V2+AF1.sub.--A2*V1)+(AF2.sub.--A1*H2+AF2.sub.--
-A2*H1)}/2 (5)
[0057] The focus detection pixel value AF_B of the combined focus
detection pixel (B image) 402 is calculated using the focus
detection pixel values AF1_B1, AF1_B2, AF2_B1, and AF2_B2 of
neighboring focus detection pixels (B images):
AF.sub.--B={(AF1.sub.--B1*V2+AF1.sub.--B2*V1)+(AF2.sub.--B1*H2+AF2.sub.--
-B2*H1)}/2 (6)
[0058] H1 and H2 represent the ratio of the horizontal distance
between AF2_A1 and AF2_A2, and that of the horizontal distance
between AF2_B1 and AF2_B2. H1 and H2 satisfy H1+H2=1.
[0059] V1 and V2 represent the ratio of the vertical distance
between AF1_A1 and AF1_A2, and that of the vertical distance
between AF1_B1 and AF1_B2. V1 and V2 satisfy V1+V2=1.
[0060] In this manner, interpolation is executed at positions where
no focus detection pixel value is obtained.
[0061] The methods of increasing the resolution by increasing the
number of focus detection pixels from images obtained by two image
sensing operations have been described. When combining AF extracted
images obtained by two or more image sensing operations, the focus
detection pixel combining circuit 121 stores a focus detection
pixel value obtained by each image sensing operation in order to
perform interpolation at positions where no focus detection pixel
value is obtained after two or more image sensing operations.
[0062] The method of increasing the resolution by increasing the
number of focus detection pixels from images obtained by two image
sensing operations is also applicable to a case wherein three or
four image sensing operations are executed. More specifically, the
number of focus detection pixels is increased from images obtained
by the first and second image sensing operations, those obtained by
the second and third image sensing operations, and those obtained
by the third and fourth image sensing operations.
[0063] As described above, the focus detection pixel combining
circuit 121 increases the number of focus detection pixels, thereby
increasing the resolution and reducing aliases. In this case,
increasing the resolution means increasing the number of SAF pixels
when calculating the defocus amount.
[0064] Information of a focus detection pixel (A image) 208 and
focus detection pixel (B image) 209 obtained by averaging images
sensed by a plurality of image sensing operations is output to the
switching circuit 123, and then to the phase difference AF circuit
108. The phase difference AF circuit 108 obtains the defocus amount
by comparing the phase differences between A and B images using the
focus detection pixels (A images) 202 and focus detection pixels (B
images) 203 before combination, and the focus detection pixel (A
image) AF_A and focus detection pixel (B image) AF_B. The phase
difference AF circuit 108 outputs the obtained defocus amount as
focusing information to the system controller 113.
[0065] In accordance with the defocus amount obtained in the phase
difference AF circuit 108, the system controller 113 notifies the
lens driving circuit 102 of the driving amount of a focus
adjustment lens in the imaging lens 101, thereby adjusting the
focus.
[0066] <Image Recording Processing>
[0067] The system controller 113 repetitively executes the
above-described focus adjustment. The signal processing circuit 109
performs signal processing such as color conversion processing for
obtained image data. The memory control circuit 110 temporarily
writes the image data in the DRAM 111.
[0068] The memory control circuit 110 reads out the image data from
the DRAM 111. The resizing circuit 112 resizes the image data to a
size suitable for display on the monitor 115 or a size suitable for
recording on the medium card 118. The memory control circuit 110
writes the resized image data again in the DRAM 111.
[0069] The memory control circuit 110 reads out the image data from
the DRAM 111. According to the NTSC or PAL scheme, the video
modulation circuit 114 modulates the image data resized to a size
suitable for display on the monitor 115. The monitor 115 displays
the image data.
[0070] The image data resized to a size suitable for recoding on
the medium card 118 is written in the medium card 118 by a card
control circuit.
[0071] <Case Wherein Object is Moving>
[0072] A matching calculation method when the object is moving will
be explained.
[0073] As shown in FIG. 5, an AF frame 802 is divided into nine,
and object motion vectors in the nine AF frames 802 are different
from each other. The motion is detected in each AF frame 802 to
detect an object distance. An AF frame 802 to be focused may be
determined automatically on the basis of the object distance
detection result of each AF frame 802 or by a user operation. When
AF evaluation is performed by tracking an object to be focused
after determining an AF frame 802 to be focused, the position of
the entire image may also be adjusted in accordance with a motion
vector detected in the AF frame 802.
[0074] <Shift Amount>
[0075] The above shift of images may occur by pixel shift. The
pixel shift includes a processing for increasing the number of
image sensing pixels or a processing for reducing a camera
shake.
[0076] The shift amount detection circuit 119 may also obtain a
shift amount when the image sensing region 201, 204, or 207 is
shifted to increase the number of image sensing pixels. The shift
amount detection circuit 119 may also obtain a shift amount when
the image sensing region 201, 204, or 207 is shifted to reduce
camera shake.
Second Embodiment
[0077] FIG. 6 is a block diagram showing the circuit arrangement of
an image sensing apparatus according to the second embodiment of
the present invention.
[0078] The same reference numerals as those in FIG. 1 which is a
block diagram of the circuit arrangement of the first embodiment
denote the same circuits.
[0079] The second embodiment is different from the first embodiment
in that a camera shake detection circuit 124 replaces the shift
amount detection circuit 119. The camera shake detection circuit
124 detects vibrations using a gyroscope or the like, and measures
a shift amount generated by camera shake. The camera shake
detection circuit 124 outputs the shift amount to a focus detection
pixel combining circuit 121 and system controller 113.
[0080] The focus detection pixel combining circuit 121 reads out
image data from the DRAM 111. The shift amount detected by the
camera shake detection circuit 124 is set as a weighting
coefficient. Then, focus detection pixels are averaged or
interpolated, as described in the first embodiment. The averaged or
interpolated focus detection pixel information is output to a phase
difference AF circuit 108 via a switching circuit 123. The phase
difference AF circuit 108 executes AF processing.
[0081] The actuator 120 in FIGS. 1 and 6 is necessary to move the
image sensor to an arbitrary position and obtain shifted images by
a plurality of image sensing operations when no camera shake
occurs. However, when camera shake occurs, the shift amount
detection circuit 119 or camera shake detection circuit 124 can
detect the shift amount between images to obtain a weighting
coefficient used to average or interpolate AF extracted images.
Thus, the actuator 120 may not be installed.
[0082] While the present invention has been described with
reference to exemplary embodiments, it is to be understood that the
invention is not limited to the disclosed exemplary embodiments.
The scope of the following claims is to be accorded the broadest
interpretation so as to encompass all such modifications and
equivalent structures and functions.
[0083] This application claims the benefit of Japanese Patent
Application No. 2008-119057, filed Apr. 30, 2008, which is hereby
incorporated by reference herein in its entirety.
* * * * *